Brett Smith

We want software creators to use the GPL and its cousin licenses. We also know that people make mistakes in the process, or don’t even try because they’ve heard it’s "too complicated." Just as we do when we develop software, we would do well to study these failures and use them as opportunities to improve the usability of the GPL. This talk aims to start that process by identifying some known problems and considering some possible solutions. (None of these solutions are a new version of the license!)

We want software creators to use FOSS licenses. We also know people make mistakes in the process, or don't even try because they've heard it's "too complicated." Just like with software, we would do well to study these failures and use them as opportunities to improve the usability of our licenses. This talk aims to start that process by identifying known problems and considering some possible solutions.

Usability issues that will be addressed in this talk include:

• License presentation
• Where a work's license is declared
• How a work's license is declared
• How creators learn to apply a license
• License drafting

In his work as the FSF's license compliance engineer, Brett had the opportunity to investigate the licensing of many projects for information. He saw what creators do, both right and wrong. He was also deeply engaged in the drafting processes for GPLv3 and MPLv2, and was able to compare and contrast where the resulting licenses succeeded and fell short at improving usability.

At LibrePlanet, we often talk about how free software intersects with human rights and social justice issues. This talk will review recent technology choices and promotion in other activist communities, discuss how recent changes in politics and technology have motivated these changes, and use these lessons to learn how the free software community can improve its outreach to other advocacy groups. We'll also look at imminent technology developments, and consider where the free software movement can work now to affect change in the future.

Large-scale scientific computing work often runs on clusters with petabytes of attached storage and specialized networking. Arvados is a free software platform to store and analyze large data sets, emphasizing reproducibility and compatibility across deployments. It's licensed under the GNU Affero General Public License version 3, with SDKs under the Apache License 2.0. This talk will provide a technical introduction to Arvados, describe how research projects like the Personal Genome Project have used it, and suggest other applications.